Comparing Current Practice to Recommendations for the Child and Adult Care Food Program

Abstract

Background:

The federal Child and Adult Care Food Program (CACFP) assists child care centers serving low-income preschoolers and regulates the quality and quantity of food served. The aim of this study was to assess the nutritional quality of lunches served at 38 child care centers and examine how current practices compare to proposed meal pattern recommendations.

Methods:

Preschool-aged children (n = 204) were observed eating lunch in 38 CACFP-participating preschools. All foods served and consumed were measured and compared to the 2011 Institute of Medicine (IOM) recommendations to improve CACFP and the 2015 Proposed Rule issued by the USDA.

Results:

All centers provided access to all required lunch components, but not all components were served (i.e., placed on the child's plate). Vegetables were significantly less likely to be served than meat or grains. Compared with CACFP recommended portion sizes, servings of meat and grain were high, whereas milk was low. Compared with IOM recommendations, average calorie consumption was appropriate, but saturated fat, protein, and sodium intake were high and dietary fiber was low. Meals that offered children both a fruit and a vegetable led to significantly higher produce consumption than meals that offered only one fruit or one vegetable.

Conclusions:

Child care centers generally comply with current CACFP regulations, but do not provide lunches consistent with the 2011 IOM recommendations for saturated fat, protein, fiber, and sodium. Decreased use of beef and cheese and increased provision of whole grains, fruits, and vegetables are recommended.

Introduction

The national Child and Adult Care Food Program (CACFP) supports food service in child care centers that serve low-income families.¹ Participating centers must serve meals and snacks that meet specific nutrition standards, and there is evidence that this program positively influences children's diets.² Centers that participate in CACFP report serving more fruit, vegetables, milk, and meat/meat alternates and fewer snacks and sweetened beverages than non-CACFP sites,³ and studies have found that children who attend CACFP participating child care centers are more likely to consume fruits and vegetables than children attending non-CACFP centers or at home under parental care.^4–6

Although current CACFP standards ensure some degree of nutritional quality, research comparing these meals to national recommendations have found that they are still high in saturated fat⁷ and sodium,⁸ highlighting potential areas for improvement. In 2008, the USDA asked the Institute of Medicine (IOM) how to bring CACFP meals into alignment with the Dietary Guidelines for Americans,⁹ and these recommendations were published in 2011.¹⁰ The 2010 Healthy Hunger-Free Kids Act required the USDA to revise the nutrition standards of the CACFP, and new USDA recommendations were released in January 2015.¹¹

To inform successful implementation of the new CACFP rules, the present study assessed current practices by direct observation of children in a sample of CACFP-participating preschools. We identified the types of foods offered and compared the nutritional profile of the foods consumed at lunch to IOM recommendations. Findings are presented to (1) determine whether lunches met current CACFP guidelines and (2) describe current consumption compared to the IOM recommendations and the USDA's January 2015 proposed rule.

Methods

Sample

The Connecticut State Department of Education assisted in identifying all CACFP-participating centers serving more than 12 preschoolers.¹² From this sample of 221 centers, 38 sites were randomly selected and invited to participate in the study. Nine declined to participate and were replaced with an additional nine randomly selected centers, all of which agreed to participate. Because the data collection agreement prohibited collection of child-level demographic information, the Census Block group data of the centers' location is presented as a proxy (see Table 1).^6,13 Lunch consumption was assessed for an average of 5 children per center from the 38 centers (n = 204). The centers reported that the children in the study ranged from 2.5 to 5.7 years old; however, we were not permitted to obtain the age of individual children.

Table 1.

Characteristics of the Child Care Centers (n = 38) and Census Block Groups in Which the Centers Are Located

Center	Percentage
NAEYC accredited	50.0
Head Start	57.9
CACFP-reimbursed meals
Breakfast	97.4
Morning snack	15.8
Lunch	97.4
Afternoon snack	79.0

	Mean	SD
Number of students enrolled in center	64.1	55.8
Minimum enrollment age, years	2.5	1.0
Maximum enrollment age, years	5.7	2.2

Block group	Mean	SD
Household median income	$43,928	$28,437
% Residents with college degree or higher	18.1	17.7
% Hispanic	35.3	31.2
% White, non-Hispanic	39.6	36.2

NAEYC, National Association for the Education of Young Children; CACFP, Child and Adult Care Food Program; SD, standard deviation.

Procedure

Child care directors provided written consent to allow observation during school hours. As thank you gifts, $125 worth of educational materials were provided to each center and directors were given $25 gift cards. The internal review board at Yale University approved all procedures involving human subjects.

Food intake was measured through visual estimates by trained researchers of the amount served (tablespoons or pieces for food; ounces for drinks) and the amount left after the meal. Before data collection, seven raters were trained for a total of 8.5 hours to visually assess food intake in the child care setting using a method based on the work of Ball and colleagues.¹⁴ All raters held master's degrees or were graduate students and had previous training in public health nutrition. To determine rater validity, an assessment of 23 common foods and beverages was conducted before data collection. One-sample t-tests were used to assess significant differences between the mean estimated portion by the raters and the true measured food portion. Of the 23 items tested, 13 were assessed perfectly (all seven raters reported an identical amount as the measured food portion) and 96% of the items were measured accurately (i.e., the null hypothesis was not rejected). Reliability among raters on the assessment was determined by an intraclass correlation (ICC) with a two-way mixed model with absolute agreement and single measures (ICC = 0.87). Raters were assessed twice during the time period of data collections to ensure that estimation did not drift.

Two or three researchers visited each center. The researchers scheduled visits in coordination with center director. Before lunch, six chairs at two or three tables were randomly chosen for observation so that researchers did not know which children would be observed. Children were not identified by name, although gender was noted. Consistent with family style feeding with this age group, children served themselves to the best of their ability and at times received assistance from the caregiver at the table. Each researcher independently recorded all foods served and consumed by 3 children sitting in the preselected seats. At each visit, raters coded the food served and consumed by 1 common child to assess inter-rater reliability. Inter-rater reliability for the amount served and consumed during the center visits remained high (ICC = 0.89).

Nutrient Assessment

To facilitate accurate nutrition analysis, researchers visited the kitchen and spoke with food preparation staff to obtain brand information, as well as cooking methods, including added fat and salt. The serving size of each food served and consumed was entered into Food Processor SQL (version 10.5.0; ESHA Research, Salem, MA) to assess: energy; protein; fat (total, saturated, and trans fat); carbohydrate; fiber; sugar; vitamins A, D, and E; the B vitamins; calcium; iron; potassium; and sodium. When brand information was missing, average nutrient profiles from the USDA database were used. All servings were converted into ounces for meat/meat alternates and grains, and cups for milk and produce.

Statistical Analyses

Linear mixed models were used to account for children nested within centers. Child care center was entered as a random effect and all other variables were entered as fixed effects. A series of z-tests for difference in proportions were used to evaluate the comparative likelihood of serving each required lunch component to each child. Because analyses were not adjusted for multiple comparisons, the statistically significant threshold was set at p < 0.01; however, a value of p < 0.05 is noted to identify trends in gender differences. Last, one-sample t-tests were used to test for differences between the average amounts of a nutrient consumed compared to the IOM recommendation for each nutrient. All analyses were performed using SPSS (version 18.0.0, 2009; SPSS, Inc., Chicago, IL) or STATA software (version 10.0, 2008; StataCorp LP, College Station, TX).

Results

Observation of Types of Foods Served and Consumed

Figure 1 illustrates the frequency of specific foods served within each category of the CACFP meal pattern. Table 2 presents the means and standard errors (adjusted for child care center as a random effect) of lunch components served and consumed. It is important to note that CACFP requirements apply to children ages 3 through 5, and the 2011 IOM recommendations are for children ages 2 through 4. Because it was not possible to exclude the 5-year-olds from the IOM comparisons, or exclude the 2-year-olds from the CACFP comparisons, Table 2 includes the portions served and consumed by our entire sample compared to both the CACFP and IOM recommendations.^9,10

Figure 1.

Most frequently served foods across centers (N = 38). *Totals exceed if centers served more than one type of the meal component. CACFP, the Child and Adult Care Food Program.

Table 2.

Amount of Each Meal Pattern Component Required at Lunch, and Adjusted Means Served and Consumed at Lunch by Gender

				CACFP study
				Boys^a		Girls
				Mean (Standard Error)
Meal component	CACFP serving size requirements ages 3–5 (current)	IOM serving size recommendations ages 2–4 (2011)	USDA proposed serving sizes ages 3–5 (2015)	Served	Consumed	Served	Consumed
Meat/meat alternate	1.5 oz	1 oz equivalent^b	1.5 oz	2.32 (0.26) oz	1.84 (0.22) oz	2.12 (0.27) oz	1.36 (0.22) oz
Grain	0.5 oz equivalent	1 oz equivalent^c	0.5 oz equivalent	1.61 (0.33) oz equivalent	1.18 (0.26) oz equivalent	1.52 (0.33) oz equivalent	0.89 (0.26) oz equivalent
Milk	0.75 cups	0.5 cups	0.75 cups	0.68 (0.05) cups	0.54 (0.04) cups	0.62 (0.05) cups	0.49 (0.04) cups
Produce (fruit/vegetable)	0.5 cup total (0.5 fruit or 0.5 vegetable)	1 cup total (0.5 fruit and 0.5 vegetable)	0.5 cup total (0.5 fruit or 0.5 vegetable)	0.41 (0.13) cups	0.23 (0.10) cups	0.47 (0.13) cups	0.22 (0.11) cups
	or		or
	(0.25 fruit and 0.25 vegetable)		(0.25 fruit and 0.25 vegetable)	0.66 (0.07) cups	0.52 (0.06) cups	0.80 (0.07) cups	0.61 (0.06) cups

(n = 102) for boys, (n = 102) for girls for Meat/meat alternate, Grain, and Milk; (n = 24) for boys, (n = 22) for girls for Fruit/Vegetable (one served); (n = 78) for boys, (n = 80) for girls for Fruit/Vegetable (both served).

According to the USDA, generally 1 ounce of meat, poultry or fish, ¼ cup cooked beans, 1 egg, 1 tablespoon of peanut buPer, or ½ ounce of nuts or seeds can be considered as 1 ounce equivalent from the Protein Foods Group.

According to the USDA, 1 slice of bread, 1 cup of ready-to-eat cereal, or ½ cup of cooked rice, cooked pasta, or cooked cereal can be considered as 1 ounce equivalent from the Grains Food Group.

IOM, Institute of Medicine; CACFP, Child and Adult Care Food Program.

Fruit and vegetables

Current CACFP standards require centers to serve either ¼ cup of both a fruit and a vegetable or ½ cup of only fruit or only vegetables. In contrast, the IOM recommends dividing fruit and vegetables into two separate categories and suggests that children receive one serving of fruit (½ cup) and two different types of vegetables (¼ cup each), totaling one cup of produce. The USDA-proposed rule adopts the IOM recommendation to divide fruits and vegetables into separate meal components, but does not increase the serving sizes. The IOM further suggests rotating a variety of vegetables throughout the week, including legumes and dark green, orange, and starchy vegetables. The USDA does not adopt this as a requirement, but does list it as a best practice.

None of the centers met the IOM produce portion size standard; however, children at centers that served both fruit and vegetables were served, on average, more produce compared with children served just one or the other (0.73 vs. 0.44 cups; p < 0.01). Consequently, children served both fruit and vegetables consumed approximately ⅓ of a cup more produce then children served just one (0.57 vs. 0.23 cups; p < 0.01). Children at centers that served either fruit or vegetables received less than the current required CACFP portion of produce (0.44 vs. 0.5 cups) and ate less than half of the target amount. As noted in Figure 1, the centers in this sample often serve dark green, orange, and starchy vegetables, but do not commonly serve legumes. Fried vegetables and fruit juice were also fairly uncommon; 8% of centers served fried vegetables and three centers served 100% juice during lunch. The fruit served was fresh (54%) or canned (46%). The vegetables were frozen (45%), fresh (37%), or canned (18%).

Grain

Current CACFP standards require centers to serve a 0.5-ounce equivalent of grain and do not have any requirements regarding whole grains. The IOM recommends doubling the grain serving to 1 ounce at lunch and calls for at least half of the grains to be whole grain or whole-grain rich. The proposed USDA rule does not adopt the recommended increase in serving sizes for grains, but does adopt the whole grain/whole grain–rich requirement for one serving per day. In our sample, girls consumed less grain than boys (0.89 vs. 1.18 ounces; p < 0.05); however, both boys and girls consumed more than the CACFP-required serving of 0.5 ounces, with boys reaching the IOM recommendation of 1.0 ounce. Only two centers served 100% whole grain bread, but staff from approximately half of the centers thought they were serving 100% whole grain because the bread was not visibly white. Only white pasta and rice were observed.

Milk

The CACFP currently requires ¾ of a cup of milk and the IOM recommends decreasing the serving to ½ a cup. The proposed rule did not decrease the milk serving size, so it remains ¾ of a cup. The IOM recommends only 1% and nonfat plain milk for children ages 2 and older. In our sample, all of the centers served milk, but the average amount served was approximately ⅔ of a cup, and children consumed an average of half a cup. Nearly all of the centers in this sample served 1% unflavored milk; six served 2% unflavored and two served flavored milk.

Meat and meat alternates

The current CACFP regulations require 1.5 ounces of meat, cheese, or an alternate protein product at lunch. The IOM recommends requiring only 1-ounce servings and recommends serving lean proteins, tofu, and legumes and limiting processed meat to once a week. The proposed rule retains the 1.5-ounce serving size and agrees with the recommendation regarding lean meats, tofu, and legumes. The USDA does not require a limit on processed meats to once a week, but recommends limiting processed meat as a best practice.

The serving sizes for meat and cheese observed exceeded the CACFP requirement and more than doubled the IOM recommendation, with boys and girls receiving an average of 2.2 ounces. Boys consumed more meat/meat alternate than girls (1.84 vs. 1.36 ounces; p < 0.05). Lean meat/meat alternates were uncommon. Regular ground beef was the most frequently served meat, appearing in nearly half of the lunches observed in the form of hamburgers, meatloaf, tacos, meatballs, or meat sauce. The next most popular meat/meat alternate was full-fat cheese, which was frequently served in combination with meat (e.g., tacos, sandwiches, and beef pasta dishes). Cheese was also popular as the only meat alternate in the meal (e.g., macaroni and cheese and pizza). Deli-style meats made up 20% of the meat/meat alternate component in the form of turkey, ham, and roast beef slices. Other reimbursable meat alternates, such as eggs, beans, nut and seed butters, and yogurt, were virtually absent from all observed lunches.

Nutrient Profile of Lunches Consumed

Table 3 presents the mean and standard error of macro and micronutrients in the lunches consumed. These are compared to the IOM lunch/supper nutrient targets for children ages 2–4, which were calculated by averaging daily nutrient targets by meal over a 5-day week.¹⁰ Overall, children consumed an appropriate average number of calories (352 kcal). The average intake of protein was 83% more than the IOM recommendation (p < 0.01). The percent of calories from fat was at the high end of the 25–35% recommendation. The percent of calories from saturated fat was 13%, which is higher than the recommended limit of 10%. The average intake of dietary fiber was 49% lower than recommended (p < 0.01).

Table 3.

Macronutrient and Micronutrient Consumption by Preschool Children From 38 Centers Compared to Institute of Medicine Recommendations

	CACFP study		IOM 2011 recommendations (lunch/supper)
Nutrient	Girls and boys (n = 204), ages 2.5–5.7		Girls and boys, ages 2–4	Percent difference between consumed and recommended (%)
Macronutrients
	Mean	±SE
Energy (kcal)^a	352	13.29	338	4.0
Fat (g)	13.6	0.7	—	—
Saturated fat (g)	5.2	0.29	—	—
Trans fat (g)	0.1	0.01	—	—
Carbohydrate (g)	39.6	1.41	—	—
Sugar (g)	17.3	0.72	—	—
Protein (g)	18.3	0.88	10	83.0^b
Dietary fiber (g)	2.8	0.13	5.5	(49.1)^b
As percentage of total energy
Fat (%)	35	0.48	25–35	—
Saturated fat (%)	13	0.19	<10	—
Carbohydrate (%	45	0.42	—	—
Protein (%)	21	0.26	—	—
Micronutrients
Vitamin C (mg)	12.4	1.2	20	(38.0)^b
Vitamin E (mg αT)	0.8	0.1	1.8	(55.6)^b
Vitamin B₆ (mg)	0.2	0.0	0.25	(20.0)
Folate (μg DFE)	49.6	2.7	82	(39.5)^b
Vitamin B₁₂ (μg)	1.2	0.1	0.72	66.7^b
Calcium (mg)	251	11.5	156	60.9^b
Phosphorous (mg)	268	11.4	217	23.5^b
Magnesium (mg)	41.2	1.6	39	5.6
Vitamin D (μg)	1.6	0.1	—	—
Vitamin A (μg RAE)	138.2	10.5	120	15.2
Vitamin K (μg)	16.9	2.7	—	—
Iron (mg)	2.1	0.1	2.2	(4.5)
Zinc (mg)	2.2	0.2	1.6	37.5^b
Sodium (mg)	600	27.8	<425	41.2^b
Potassium (mg)	484.5	19.4	849	(42.9)^b

Numbers in parentheses represent that less of the nutrient was consumed than recommended by the IOM; numbers without parentheses represent that more of the nutrient was consumed than recommended.

1 kcal_th = 4.184 kJ.

Denotes a statistically significant difference for a one-sample t-test at the p < 0.01 level.

CACFP, Child and Adult Care Food Program; IOM, Institute of Medicine; αT, α-tocopherol; DFE, dietary folate equivalent; RAE, retinol activity equivalent; SE, standard error.

Children's consumption of many micronutrients was at or above what would be expected from one lunch. Notably, calcium intake was adequate, as were vitamin B12, vitamin A, phosphorous, magnesium, and zinc. In contrast, vitamin B6, vitamin C, vitamin E, folate, and potassium were low. Iron was just under recommended levels (2.1 vs. 2.2 mg; p = 0.55). Sodium intake was high (600 vs. <425 mg), and further analysis of the source of sodium revealed that meat/meat alternates (such as seasoned beef in tacos, hamburgers, meatballs, beef stroganoff, and meatloaf) and mixed dishes containing meat/meat alternates (such as prepared pizza and macaroni and cheese) provided 47% of the sodium in the meal. Other sources of sodium were grains (18%), milk (15%), and condiments (8%).

Lunch Components Available Versus Served

The majority of centers served lunch family style, where caregivers ate at the table with the children and food was served by a combination of adult-assisted and child self-serving. The adults provided assistance as much or as little as they perceived the child needed, and there was variability in how encouraging caregivers were for children to take all of the food components. As a result, even though all centers offered the full complement of reimbursable lunch components, every component was not necessarily served to each child.

To examine whether the children and caregivers favored certain meal components, z-tests for difference in proportions were used to compare the proportion of children who had different components on their plates during the meal. The proportion of children with vegetables on their plates (when vegetables were part of the lunch; n = 189) was compared to the proportion of children with grains on their plates. Twenty-nine children (14.2%) were not served vegetables, whereas only 6 (2.9%) were not served grains (z = 4.31; p < 0.001). Vegetables were also less likely to be served than meat/meat alternates, which were served to all but 17 children (8.3%; z = 2.16; p = 0.031). Fruit was available in the meals for 178 children, but 20 were not served fruit (9.8%), making fruit less likely to be served than grain (2.9%; z = 3.21; p = 0.001). Likelihood of serving fruit did not differ significantly from that of serving meat/meat alternates or vegetables.

Discussion and Conclusions

Child care centers generally comply with current CACFP regulations, but do not provide lunches consistent with the 2011 IOM recommendations. The nutrient analysis indicated that, on average, meals provided children with appropriate amounts of energy, but highlighted the overconsumption of saturated fat and sodium, and the percent of energy from saturated fat (13%) exceeded the 10% recommended by the IOM. These findings are consistent with those reported in similar studies by Oakley and colleagues and Maalouf and colleagues.^7,15

The observed high levels of saturated fat and sodium were linked to the meat/meat alternates served. Beef dominated the meals, and though it is an excellent source of protein and iron, high amounts of red meat have also been associated with an increased risk of cancer in adults.¹⁶ Further, the American Heart Association's (AHA) guidelines for the general population (including children), recommend limiting saturated fats and salt to achieve desirable blood pressure and decreased risk of cardiovascular disease.¹⁷ The AHA advocates switching to lean meats and nonmeat alternates, such as legumes.

The present study also provides insight into how current practices compare to the IOM recommendations and the USDA proposed rule and highlights potential challenges in meeting new standards. Introducing more whole grain products will require significant changes in the products served. Only two centers served 100% whole grain bread, but staff from approximately half of the centers thought they were serving 100% whole grain because the bread was not visibly white. Based on the current data and previous research, accurately identifying whole grains can be a challenge.^18,19 Training on how to identify 100% whole grain and whole grain–rich products is critical for successful implementation of this policy.

Children in this sample were served only ½ cup of milk on average, even though the minimum has been ¾ of a cup. Further research and training is needed to understand why milk is consistently underserved and develop strategies to ensure that children receive the recommended amount of milk. We observed that many of the cups used by the children were very small, making it impossible to serve ¾ of a cup at once. Staff also appeared concerned about spills, which may have inadvertently led to them providing smaller portions.

One key change that is included in the proposed rule is to consider fruits and vegetables two different meal components, even though the required portion sizes were not changed as recommended by the IOM. The current data support the effectiveness of that policy change because when it occurred naturally among the centers in the study, children ate more produce overall. As a note, larger portions of produce were served to the children when both a fruit and a vegetable were available, so the increased consumption may be owing to more variety, larger portions, or both. There is evidence from previous studies that increasing variety can increase consumption,^20,21 and increasing portion size may also lead to greater consumption among older preschoolers.²² When this policy change goes into effect nationally, it will be interesting to test whether increasing variety also increases the portion size served and consumed.

Family-style feeding continues to be the recommended strategy for preschoolers because it allows children to practice their motor skills, learn about dining norms, and regulate the type and amount of food on their plates.²³ The risk in allowing children complete autonomy in serving themselves is that some children may not try new foods. A challenge identified in the present study was ensuring that fruit and vegetables were served to each child. Our finding that vegetables and fruit were the least likely to be served suggests that some children did not want them served. Previous research with elementary school–aged children found that a prompt to take the fruit serving at lunch resulted in greater fruit intake.²⁴ Very young children, however, may become upset at having unwanted foods on their plates. Refining family style to include having teachers routinely serve or encourage the child to take a small portion of each component may change the social norm within the child care setting and lead to better development of a diverse palate.

When evaluating this study, it is important to note we only have data from 1 day at each center within a single state, so our findings may not reflect the range of practices at each center and may differ from practices in other states. Further, our study may present an overly positive picture of meal quality because more than half of our centers participated in Head Start, which has its own set of nutrition requirements.³ Finally, caregivers and children may have adjusted their behavior because of our presence. Even with these caveats, these findings provide a valuable opportunity to compare current practices to both the IOM recommendations and the USDA-proposed rule and identify the types of technical assistance that will be needed to successfully implement new CACFP policies.

Footnotes

Acknowledgment

This project was funded by the Robert Wood Johnson Foundation Healthy Eating Research program. We are grateful to the Yale University students who assisted with data collection.

Author Disclosure Statement

No competing financial interests exist.

References

United States Department of Agriculture Food and Nutrition Service. Food and Nutrition Service. Why CACFP is Important?. 2013. Available at www.fns.usda.gov/cacfp/why-cacfp-important Last accessed March 25, 2015.

United States Department of Agriculture Food and Nutrition Service. Child and Adult Care Food Program: Meal patterns.. 2010. Available at www.fns.usda.gov/cnd/care/ProgramBasics/Meals/Meal_Patterns.htm Last accessed August 25, 2010.

Ritchie

, Boyle

, Chandran

, et al. Participation in the child and adult care food program is associated with more nutritious foods and beverages in child care. Child Obes, 2012; 8:224–229.

Gordon

, Kaestner

, Korenman

, et al. The Child and Adult Care Food Program: Who is served and what are their nutritional outcomes?. National Bureau of Economic Research: Cambridge, MA, 2010.

Cepinsek

, Burstein

. Maternal employment and children's nutrition: Vol. II. Other nutrition-related outcomes (E-FAN-04-006-2). Economic Research Service, United States Department of Agriculture: Washington, DC, 2004.

US Census Bureau. American Community Survey. American Community Survey 1-year estimates.. 2008. Available at http://factfinder.census.gov Last accessed February 8, 2015.

Oakley

, Bomba

, Knight

, et al. Evaluation of menus planned in Mississippi child-care centers participating in the Child and Adult Care Food Program. J Am Diet Assoc, 1995; 95:765–768.

Fox

, Glantz

, Geitz

, et al. Early Childhood and Child Care Study: Nutritional Assessment of the CACFP. Abt Associates Inc.: Alexandria, VA, 1997.

United States Department of Agriculture Food and Nutrition Service. Child and Adult Care Food Program, meals and snacks. Child care meal pattern (p. 2). United States Department of Agriculture: Washington, DC, 2013.

10.

Institute of Medicine. Child and Adult Care Food Program: Aligning dietary guidance for all. The National Academies Press: Washington, DC, 2011, Table 7-3, p. 120; Table E-125, p. 216.

11.

USDA Food and Nutrition Service. Healthy, Hunger-free Kids Act of 2010. January 2015. Available at www.federalregister.gov/articles/2015/01/15/2015-00446/child-and-adult-care-food-program-meal-pattern-revisions-related-to-the-healthy-hunger-free-kids-act Last accessed April 10, 2015.

12.

Fiore

. Final report for Connecticut's 2007–2010 Team Nutrition Grant. Connecticut State Department of Education, Bureau of Health/Nutrition, Family Services and Adult Education: Hartford, CT, November 15, 2010.

13.

Henderson

, Grode

, Middleton

, et al. Validity of a measure to assess the child care nutrition and physical activity environment. J Am Diet Assoc., 2011; 111:1306–1313.

14.

Ball

, Benjamin

, Ward

. Development and reliability of an observation method to assess food intake of young children in child care. J Am Diet Assoc, 2007; 107:656–661.

15.

Maalouf

, Evers

, Griffin

, et al. Assessment of mealtime environments and nutrition practices in child care centers in Georgia. Child Obes, 2013; 9:437–445.

16.

Pan

, Sun

, Bernstein

, et al. Red meat consumption and mortality: Results from 2 prospective cohort studies. Arch Intern Med, 2012; 172:555–563.

17.

American Heart Association, Gidding

, Dennison

, et al. Dietary recommendations for children and adolescents: A guide for practioners. Pediatrics, 2006; 117:544–559.

18.

Burgess-Champoux

, Marquart

, Vickers

, et al. Perceptions of children, parents, and teachers regarding whole-grain foods, and implications for a school-based intervention. J Nutr Educ Behav, 2006; 38:230–237.

19.

Adams

, Engstrom

. Helping consumers achieve recommended intakes of whole grain foods. J Am Coll Nutr, 2000; 19:339S–344S.

20.

Rolls

. Sensory-specific satiety. Nutr Rev, 1986; 44:93–104.

21.

Roe

, Meengs

, Birch

, et al. Serving a variety of vegetables and fruit as a snack increased intake in preschool children. Am J Clin Nutr, 2013; 98:693–699.

22.

Rolls

, Engell

, Birch

. Serving portion size influences 5-year-old but not 3-year-old children's food intakes. J Am Diet Assoc, 2000; 100:232–234.

23.

American Academy of Pediatrics, American Public Health Association, and National Resource Center for Health and Safety in Child Care and Early Education. 2012. Preventing childhood obesity in early care and education: Selected standards from Caring for our children: National health and safety performance standards: Guidelines for early care and education programs, 3rd edition. Available at nrckids.org/CFOC3/PDFVersion/preventing_obesity.pdf Last accessed September 10, 2015.

24.

Schwartz

. The influence of a verbal prompt on school lunch fruit consumption: A pilot study. Int J Behav Nutr Phys Act, 2007; 4:6.